For several days now, the echoes of the loud prank on President Duda, who instead of President Macron, talked to Russian pranksters – Vladimir Kuznetsov (Vovan) and Alexei Stoljarov (Lexus) – have not been silent.
As part of our current research and development activity, we biometrically analyzed the recordings of the pranksters’ voices and compared them with the voice of the real Macron (Polish and English versions). We downloaded all voice samples in the form of individual recordings from the public domain on YouTube. Our goal was to confirm the effectiveness of biometric systems for this specific situation – identifying fraud.
What did the BiometrIQ analysis show? It turned out that the voice of one of the “Lexus” pranksters was just over 50% consistent with the voice of the President of France and as much as 97% consistent with the voice of the false president. The voice of the second one – “Vovana” – showed no similarities (0%) to the fake president.
This clearly proves that thanks to biometric analysis we managed to:
=> detect the fact, only after 1 minute, that a fake president was involved in the conversation
=> identify the identity of the fictional president (Lexus)
=> confirm that the public domain is a very good source of voice samples, which may not always be used for noble purposes
=> strengthen the thesis that attacks using social engineering are the most effective, and in this case it was the choice of the right time when we are dealing with increased stress (rocket drop).
Over 600 people, 20 panel sessions, 3 thematic paths are the balance of this year’s Digital Summit, which took place in Lublin. The event was attended by representatives of government administration, foundations as well as business, research centers and schools. The Forum could not miss the BiometrIQ company, which has been specializing in development of voice technology systems and their use in fraud detection for several years.
Being an organizer of one of panels, the company introduced the voice biometrics market by inviting experts representing the technology, business and science sectors to the discussion “Stop voice theft. Innovative voice biometric services”. The invited guests discussed issues that are extremely important for this field, such as the rapid pace of development of voice technologies and the associated threats, as well as the range of applications of voice biometrics and ways to protect voice theft in the network, thus outlining the near future for the voice market.
The panel session began with a fragment of popular film “Mission Impossible 3” showing an example of voice theft and its effective use.
Moderator Maciej Kurzajewski, referring to the film, opened the discussion with the words “Is what we saw a moment ago on this screen already reality or the near future or was the cinema one step ahead of what is happening today”?
Adrian Boczkowski, the owner of BQ Consulting, spoke about readiness of business to use biometric technology. He emphasized that commercial solutions are currently mainly interested in financial institutions and telecoms, and in the future they will also be the government and health sectors. In fact, it cannot be ruled out that voice biometrics will eventually appear in every aspect of our lives. This is primarily due to the fact that speech is the most natural mechanism of communication. It is estimated, Adrian Boczkowski continued, that the global voice biometrics market will increase even 3 times by 2024.
The driver of effective voice synthesis will be artificial intelligence and neural networks. According to Tomasz Jadczyk, CTO Techmo, this technology will play a key role in developing voice replication. It will allow for building more advanced user models, thanks to which future solutions will be even more perfect than before. Already now to create a good quality model of the user’s voice, it is enough from several minutes to an hour of recording, while just a few years ago several hours of recordings were needed.
Mentioned naturalness and comfort of voice communication mean that the importance of voice technologies will grow. Referring to the available research, Adrian Boczkowski said that 50% of the population is convinced to use this authentication method, and nearly 80% consider it a better method than a pin or password. This may mean that soon voice login will become a part of everyday life. However, we must remember existing threats posed by voice biometrics. “It is very easy to steal a voice today,” said Andrzej Tymecki, managing director of BiometrIQ. Millions of recordings are circulating on the web, which, due to growing popularity of social media, we willingly leave ourselves. Citing numbers, Andrzej Tymecki pointed out that as of today, the YouTube archives alone contain 800 million recordings with an average length of 11.7 minutes, which gives about 70 seconds of recording assigned to every person living on the Earth. It is these recordings left in the public domain that can be easily used to create a replica of someone’s voice, which can be used to carry out social engineering attacks. In particular, public persons: journalists, politicians and people holding important positions in companies are at risk of voice theft in order to use it in such attacks.
Recommending caution in communicating with bots seems to be a good tactic. Michał Hałoń, an expert at NASK-PIB, warned against the dangers associated with, among others, uncritical communication with voice assistants, which may result in leaving voice samples in an uncontrolled way, often in combination with personal data provided by us during the conversation. Among the most popular methods of voice theft protection, he pointed to identity verification by callback or video call.
Share the article
Cookie policy
We use cookies
This website uses cookies. Cookies are served as text files stored by the user. They are used to store user detection information, analyze traffic, and apply site content and features. If you want more information, please read cookie policy
These cookies are necessary for the website to function and cannot be disabled in our systems. They are usually only set in response to actions you take, which is limited to requesting services, such as setting privacy options, filling out forms. You can set your browser to block or warn you about these cookies, but some parts of the website will not function properly. These cookies do not store any personal data.
These cookies allow us to count visits and traffic sources so that we can measure and improve the performance of our website. They help us learn which pages are most and least popular and see how visitors move around the site. All information collected by these cookies is aggregated and anonymous. If you do not allow these cookies, we will not know when you have visited our website and will not be able to monitor its performance.